Skip to content

Can Mac M4 Max 36G run Gemma 4 12B IT?

Comfortable: GGUF Q8_0 needs 13.8 GB at 4K, batch 1; headroom 13.2 GB.

Text input, FP16 KV cache, one card. Unified-memory capacity uses the GPU guide’s usable limit; negative headroom may require host offload. Speed source settings may differ from this 4K estimate.

Measured runs

No measured run for this pair. Sizing is an estimate, not a runtime test.

Quant memory at 4K

QuantTotal GBVerdictWeight sizing
GGUF Q4_K_M8.0ComfortableVerified file
GGUF Q8_013.8ComfortableVerified file

Context capacity · GGUF Q8_0

ContextTotal GBVerdict
4K13.8Comfortable
16K14.0Comfortable
32K14.3Comfortable
128K15.9Comfortable

Generation speed

GGUF Q8_0: 21–28 tok/s · Derived range

Other model / GPU pairs